[PATCH 0/2] AF_PACKET fanout support

STALE5508d

8 messages, 3 authors, 2011-07-06 · open the first message on its own page

[PATCH 0/2] AF_PACKET fanout support

From: David Miller <davem@davemloft.net>
Date: 2011-07-05 04:20:11

This is a fully functional version, I've tested both hash and
load-balance modes successfully.  I plan to commit this to
net-next-2.6 very soon.

Below is a test program that other people can play with
if they want.  It basically creates 4 threads, and creates
an AF_PACKET fanout amongst them.  Each thread prints out
it's pid in parentheses every time it receives 10 packets.
After each thread processes 10,000 packets, it exits.

Try things like "./test eth0 hash", "./test eth0 lb", etc.

Signed-off-by: David S. Miller <davem@davemloft.net>

--------------------
#include <stddef.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>

#include <sys/types.h>
#include <sys/wait.h>
#include <sys/socket.h>
#include <sys/ioctl.h>

#include <unistd.h>

#include <linux/if_ether.h>
#include <linux/if_packet.h>

#include <net/if.h>

static const char *device_name;
static int fanout_type;
static int fanout_id;

#ifndef PACKET_FANOUT
#define PACKET_FANOUT		18
#define PACKET_FANOUT_HASH		0
#define PACKET_FANOUT_LB		1
#endif

static int setup_socket(void)
{
	int err, fd = socket(AF_PACKET, SOCK_RAW, htons(ETH_P_IP));
	struct sockaddr_ll ll;
	struct ifreq ifr;
	int fanout_arg;

	if (fd < 0) {
		perror("socket");
		return EXIT_FAILURE;
	}

	memset(&ifr, 0, sizeof(ifr));
	strcpy(ifr.ifr_name, device_name);
	err = ioctl(fd, SIOCGIFINDEX, &ifr);
	if (err < 0) {
		perror("SIOCGIFINDEX");
		return EXIT_FAILURE;
	}

	memset(&ll, 0, sizeof(ll));
	ll.sll_family = AF_PACKET;
	ll.sll_ifindex = ifr.ifr_ifindex;
	err = bind(fd, (struct sockaddr *) &ll, sizeof(ll));
	if (err < 0) {
		perror("bind");
		return EXIT_FAILURE;
	}

	fanout_arg = (fanout_id | (fanout_type << 16));
	err = setsockopt(fd, SOL_PACKET, PACKET_FANOUT,
			 &fanout_arg, sizeof(fanout_arg));
	if (err) {
		perror("setsockopt");
		return EXIT_FAILURE;
	}

	return fd;
}

static void fanout_thread(void)
{
	int fd = setup_socket();
	int limit = 10000;

	if (fd < 0)
		exit(fd);

	while (limit-- > 0) {
		char buf[1600];
		int err;

		err = read(fd, buf, sizeof(buf));
		if (err < 0) {
			perror("read");
			exit(EXIT_FAILURE);
		}
		if ((limit % 10) == 0)
			fprintf(stdout, "(%d) \n", getpid());
	}

	fprintf(stdout, "%d: Received 10000 packets\n", getpid());

	close(fd);
	exit(0);
}

int main(int argc, char **argp)
{
	int fd, err;
	int i;

	if (argc != 3) {
		fprintf(stderr, "Usage: %s INTERFACE {hash|lb}\n", argp[0]);
		return EXIT_FAILURE;
	}

	if (!strcmp(argp[2], "hash"))
		fanout_type = PACKET_FANOUT_HASH;
	else if (!strcmp(argp[2], "lb"))
		fanout_type = PACKET_FANOUT_LB;
	else {
		fprintf(stderr, "Unknown fanout type [%s]\n", argp[2]);
		exit(EXIT_FAILURE);
	}

	device_name = argp[1];
	fanout_id = getpid() & 0xffff;

	for (i = 0; i < 4; i++) {
		pid_t pid = fork();

		switch (pid) {
		case 0:
			fanout_thread();

		case -1:
			perror("fork");
			exit(EXIT_FAILURE);
		}
	}

	for (i = 0; i < 4; i++) {
		int status;

		wait(&status);
	}

	return 0;
}

Re: [PATCH 0/2] AF_PACKET fanout support

From: Tom Herbert <hidden>
Date: 2011-07-06 00:46:41

Dave,

Thanks for these patches!  Is it possible you could use an alternative
term than "fanout"?  I think this may be more often associated with a
transmit operation (e.g. multicast fanout).

Also, another useful mode of steering would be to steer packets to a
socket which was recently processed by a thread running on the same
CPU; somewhat analogous to RFS (cc'ed WIllem Bruijn who is already
working on this I believe).

Tom

On Mon, Jul 4, 2011 at 9:20 PM, David Miller [off-list ref] wrote:
This is a fully functional version, I've tested both hash and
load-balance modes successfully.  I plan to commit this to
net-next-2.6 very soon.

Below is a test program that other people can play with
if they want.  It basically creates 4 threads, and creates
an AF_PACKET fanout amongst them.  Each thread prints out
it's pid in parentheses every time it receives 10 packets.
After each thread processes 10,000 packets, it exits.

Try things like "./test eth0 hash", "./test eth0 lb", etc.

Signed-off-by: David S. Miller <davem@davemloft.net>

--------------------
#include <stddef.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>

#include <sys/types.h>
#include <sys/wait.h>
#include <sys/socket.h>
#include <sys/ioctl.h>

#include <unistd.h>

#include <linux/if_ether.h>
#include <linux/if_packet.h>

#include <net/if.h>

static const char *device_name;
static int fanout_type;
static int fanout_id;

#ifndef PACKET_FANOUT
#define PACKET_FANOUT           18
#define PACKET_FANOUT_HASH              0
#define PACKET_FANOUT_LB                1
#endif

static int setup_socket(void)
{
       int err, fd = socket(AF_PACKET, SOCK_RAW, htons(ETH_P_IP));
       struct sockaddr_ll ll;
       struct ifreq ifr;
       int fanout_arg;

       if (fd < 0) {
               perror("socket");
               return EXIT_FAILURE;
       }

       memset(&ifr, 0, sizeof(ifr));
       strcpy(ifr.ifr_name, device_name);
       err = ioctl(fd, SIOCGIFINDEX, &ifr);
       if (err < 0) {
               perror("SIOCGIFINDEX");
               return EXIT_FAILURE;
       }

       memset(&ll, 0, sizeof(ll));
       ll.sll_family = AF_PACKET;
       ll.sll_ifindex = ifr.ifr_ifindex;
       err = bind(fd, (struct sockaddr *) &ll, sizeof(ll));
       if (err < 0) {
               perror("bind");
               return EXIT_FAILURE;
       }

       fanout_arg = (fanout_id | (fanout_type << 16));
       err = setsockopt(fd, SOL_PACKET, PACKET_FANOUT,
                        &fanout_arg, sizeof(fanout_arg));
       if (err) {
               perror("setsockopt");
               return EXIT_FAILURE;
       }

       return fd;
}

static void fanout_thread(void)
{
       int fd = setup_socket();
       int limit = 10000;

       if (fd < 0)
               exit(fd);

       while (limit-- > 0) {
               char buf[1600];
               int err;

               err = read(fd, buf, sizeof(buf));
               if (err < 0) {
                       perror("read");
                       exit(EXIT_FAILURE);
               }
               if ((limit % 10) == 0)
                       fprintf(stdout, "(%d) \n", getpid());
       }

       fprintf(stdout, "%d: Received 10000 packets\n", getpid());

       close(fd);
       exit(0);
}

int main(int argc, char **argp)
{
       int fd, err;
       int i;

       if (argc != 3) {
               fprintf(stderr, "Usage: %s INTERFACE {hash|lb}\n", argp[0]);
               return EXIT_FAILURE;
       }

       if (!strcmp(argp[2], "hash"))
               fanout_type = PACKET_FANOUT_HASH;
       else if (!strcmp(argp[2], "lb"))
               fanout_type = PACKET_FANOUT_LB;
       else {
               fprintf(stderr, "Unknown fanout type [%s]\n", argp[2]);
               exit(EXIT_FAILURE);
       }

       device_name = argp[1];
       fanout_id = getpid() & 0xffff;

       for (i = 0; i < 4; i++) {
               pid_t pid = fork();

               switch (pid) {
               case 0:
                       fanout_thread();

               case -1:
                       perror("fork");
                       exit(EXIT_FAILURE);
               }
       }

       for (i = 0; i < 4; i++) {
               int status;

               wait(&status);
       }

       return 0;
}
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 0/2] AF_PACKET fanout support

From: David Miller <davem@davemloft.net>
Date: 2011-07-06 01:20:49

From: Tom Herbert <redacted>
Date: Tue, 5 Jul 2011 17:46:36 -0700
Thanks for these patches!  Is it possible you could use an alternative
term than "fanout"?  I think this may be more often associated with a
transmit operation (e.g. multicast fanout).
I've never heard such terminology myself.

Sorry, the fanout name is staying :-)
Also, another useful mode of steering would be to steer packets to a
socket which was recently processed by a thread running on the same
CPU; somewhat analogous to RFS (cc'ed WIllem Bruijn who is already
working on this I believe).
This sounds like a good way to overload a local socket and prevent
pushing the work to lesser used sockets on other cpus.

Re: [PATCH 0/2] AF_PACKET fanout support

From: Tom Herbert <hidden>
Date: 2011-07-06 03:13:30

quoted
Also, another useful mode of steering would be to steer packets to a
socket which was recently processed by a thread running on the same
CPU; somewhat analogous to RFS (cc'ed WIllem Bruijn who is already
working on this I believe).
This sounds like a good way to overload a local socket and prevent
pushing the work to lesser used sockets on other cpus.
Sure, it you're not using RPS or RSS!  These should already be
distributing the RX work amongst CPUs.

Re: [PATCH 0/2] AF_PACKET fanout support

From: David Miller <davem@davemloft.net>
Date: 2011-07-06 03:20:00

From: Tom Herbert <redacted>
Date: Tue, 5 Jul 2011 20:13:27 -0700
quoted
quoted
Also, another useful mode of steering would be to steer packets to a
socket which was recently processed by a thread running on the same
CPU; somewhat analogous to RFS (cc'ed WIllem Bruijn who is already
working on this I believe).
This sounds like a good way to overload a local socket and prevent
pushing the work to lesser used sockets on other cpus.
Sure, it you're not using RPS or RSS!  These should already be
distributing the RX work amongst CPUs.
One idea I did have while working on the PACKET_FANOUT bits was
to allow a packet socket to be bound to a particular cpu.  And
to implement this we'd have a per-cpu list of packet_type taps.

But in order for the user to make sure he gets all the traffic,
he'd have to make sure he bound one AF_PACKET socket to every
online cpu and then listened for all cpu hotplug events.

It doesn't really work.

Re: [PATCH 0/2] AF_PACKET fanout support

From: Eric Dumazet <hidden>
Date: 2011-07-06 04:07:52

Le mardi 05 juillet 2011 à 20:19 -0700, David Miller a écrit :
From: Tom Herbert <redacted>
Date: Tue, 5 Jul 2011 20:13:27 -0700
quoted
quoted
quoted
Also, another useful mode of steering would be to steer packets to a
socket which was recently processed by a thread running on the same
CPU; somewhat analogous to RFS (cc'ed WIllem Bruijn who is already
working on this I believe).
This sounds like a good way to overload a local socket and prevent
pushing the work to lesser used sockets on other cpus.
Sure, it you're not using RPS or RSS!  These should already be
distributing the RX work amongst CPUs.
One idea I did have while working on the PACKET_FANOUT bits was
to allow a packet socket to be bound to a particular cpu.  And
to implement this we'd have a per-cpu list of packet_type taps.

But in order for the user to make sure he gets all the traffic,
he'd have to make sure he bound one AF_PACKET socket to every
online cpu and then listened for all cpu hotplug events.

It doesn't really work.
It is working right now if you dont have too many cpus, adding as many
sockets as possible cpus, and convenient BPF filter (matching CPU X) per
packet socket. Of course, if a cpu is offlined, the corresponding socket
wont receive any packet.

Currently, with a multiqueue NIC, the two policies you have might be in
conflict with NIC flow distribution among its queues.

In the end, lot of different cpus will access all the sockets.

I suspect this can be solved adding a third policy : hash by CPU only



Re: [PATCH 0/2] AF_PACKET fanout support

From: David Miller <davem@davemloft.net>
Date: 2011-07-06 06:44:33

From: Eric Dumazet <redacted>
Date: Wed, 06 Jul 2011 06:07:46 +0200
I suspect this can be solved adding a third policy : hash by CPU only
Agreed, I'll implement this policy.

Re: [PATCH 0/2] AF_PACKET fanout support

From: David Miller <davem@davemloft.net>
Date: 2011-07-06 06:55:46

From: David Miller <davem@davemloft.net>
Date: Tue, 05 Jul 2011 23:44:24 -0700 (PDT)
From: Eric Dumazet <redacted>
Date: Wed, 06 Jul 2011 06:07:46 +0200
quoted
I suspect this can be solved adding a third policy : hash by CPU only
Agreed, I'll implement this policy.
packet: Add 'cpu' fanout policy.

Unfortunately we have to use a real modulus here as
the multiply trick won't work as effectively with cpu
numbers as it does with rxhash values.

Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/include/linux/if_packet.h b/include/linux/if_packet.h
index 84e684e..c148606 100644
--- a/include/linux/if_packet.h
+++ b/include/linux/if_packet.h
@@ -53,6 +53,7 @@ struct sockaddr_ll {
 
 #define PACKET_FANOUT_HASH		0
 #define PACKET_FANOUT_LB		1
+#define PACKET_FANOUT_CPU		2
 #define PACKET_FANOUT_FLAG_DEFRAG	0x8000
 
 struct tpacket_stats {
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 7ba6871..41f0489 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -447,6 +447,13 @@ static struct sock *fanout_demux_lb(struct packet_fanout *f, struct sk_buff *skb
 	return f->arr[cur];
 }
 
+static struct sock *fanout_demux_cpu(struct packet_fanout *f, struct sk_buff *skb, unsigned int num)
+{
+	unsigned int cpu = smp_processor_id();
+
+	return f->arr[cpu % num];
+}
+
 static struct sk_buff *fanout_check_defrag(struct sk_buff *skb)
 {
 	const struct iphdr *iph;
@@ -482,8 +489,8 @@ static struct sk_buff *fanout_check_defrag(struct sk_buff *skb)
 	return skb;
 }
 
-static int packet_rcv_fanout_hash(struct sk_buff *skb, struct net_device *dev,
-				  struct packet_type *pt, struct net_device *orig_dev)
+static int packet_rcv_fanout(struct sk_buff *skb, struct net_device *dev,
+			     struct packet_type *pt, struct net_device *orig_dev)
 {
 	struct packet_fanout *f = pt->af_packet_priv;
 	unsigned int num = f->num_members;
@@ -496,35 +503,25 @@ static int packet_rcv_fanout_hash(struct sk_buff *skb, struct net_device *dev,
 		return 0;
 	}
 
-	if (f->defrag) {
-		skb = fanout_check_defrag(skb);
-		if (!skb)
-			return 0;
-	}
-
-	skb_get_rxhash(skb);
-
-	sk = fanout_demux_hash(f, skb, num);
-	po = pkt_sk(sk);
-
-	return po->prot_hook.func(skb, dev, &po->prot_hook, orig_dev);
-}
-
-static int packet_rcv_fanout_lb(struct sk_buff *skb, struct net_device *dev,
-				struct packet_type *pt, struct net_device *orig_dev)
-{
-	struct packet_fanout *f = pt->af_packet_priv;
-	unsigned int num = f->num_members;
-	struct packet_sock *po;
-	struct sock *sk;
-
-	if (!net_eq(dev_net(dev), read_pnet(&f->net)) ||
-	    !num) {
-		kfree_skb(skb);
-		return 0;
+	switch (f->type) {
+	case PACKET_FANOUT_HASH:
+	default:
+		if (f->defrag) {
+			skb = fanout_check_defrag(skb);
+			if (!skb)
+				return 0;
+		}
+		skb_get_rxhash(skb);
+		sk = fanout_demux_hash(f, skb, num);
+		break;
+	case PACKET_FANOUT_LB:
+		sk = fanout_demux_lb(f, skb, num);
+		break;
+	case PACKET_FANOUT_CPU:
+		sk = fanout_demux_cpu(f, skb, num);
+		break;
 	}
 
-	sk = fanout_demux_lb(f, skb, num);
 	po = pkt_sk(sk);
 
 	return po->prot_hook.func(skb, dev, &po->prot_hook, orig_dev);
@@ -571,6 +568,7 @@ static int fanout_add(struct sock *sk, u16 id, u16 type_flags)
 	switch (type) {
 	case PACKET_FANOUT_HASH:
 	case PACKET_FANOUT_LB:
+	case PACKET_FANOUT_CPU:
 		break;
 	default:
 		return -EINVAL;
@@ -606,14 +604,7 @@ static int fanout_add(struct sock *sk, u16 id, u16 type_flags)
 			atomic_set(&match->sk_ref, 0);
 			match->prot_hook.type = po->prot_hook.type;
 			match->prot_hook.dev = po->prot_hook.dev;
-			switch (type) {
-			case PACKET_FANOUT_HASH:
-				match->prot_hook.func = packet_rcv_fanout_hash;
-				break;
-			case PACKET_FANOUT_LB:
-				match->prot_hook.func = packet_rcv_fanout_lb;
-				break;
-			}
+			match->prot_hook.func = packet_rcv_fanout;
 			match->prot_hook.af_packet_priv = match;
 			dev_add_pack(&match->prot_hook);
 			list_add(&match->list, &fanout_list);
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help